Scholarlink

<p>This research presents a comprehensive study on the fabrication and tribomechanical characterization of aluminum matrix composites reinforced with titanium carbide (TiC) in Aluminum 7075 (Al7075). The primary objective is to reduce wear rates and enhance mechanical performance through controlled reinforcement, and to optimize processing and testing parameters using statistical and machine learning-based techniques. Al7075/TiC composites with varying TiC content (0–4 wt%) were fabricated using the powder metallurgy route, which involved mechanical mixing, cold compaction, and sintering to achieve homogeneous particle distribution and high interfacial integrity. The fabricated composites were evaluated for their microstructural, mechanical, and tribological properties. Experiments were conducted under various applied loads (10–30&#xa0;N) and sliding speeds (1–3&#xa0;m/s) to investigate the wear behavior. Microstructural analysis confirmed uniform dispersion of TiC particles and strong interfacial bonding with the matrix. Results indicated that increasing TiC content improved composite density (up to 61.42%) and hardness (up to 86.95%). Response Surface Methodology (RSM) was applied in conjunction with advanced optimization strategies inspired by deep learning, including Grid Search, Random Search, and Bayesian Optimization, to identify the optimal parameters for minimizing wear rate. All techniques consistently identified the optimal condition as 4 wt% TiC, 10&#xa0;N load, and 3&#xa0;m/s sliding speed, yielding the lowest wear rate of 0.00214&#xa0;mm<sup>3</sup>/min. Analysis of variance revealed that TiC content had the most significant effect on wear rate, followed by applied load and sliding speed. Among the optimization methods, Bayesian Optimization demonstrated superior efficiency by effectively exploring the parameter space, making it particularly advantageous for studies constrained by limited experimental data.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Integrated Optimization and Tribo-Mechanical Evaluation of Al7075/TiC Composites Using RSM and Data-Driven Optimization Techniques

  • 期刊论文

  • 发表日期：

    2025年12月1日

查看全文
  • Mulugundam Siva Surya,
  • Votarikari Naveen Kumar,
  • Jay Prakash Srivastava,
  • S. K. Gugulothu,
  • G. Prasanthi

摘要

This research presents a comprehensive study on the fabrication and tribomechanical characterization of aluminum matrix composites reinforced with titanium carbide (TiC) in Aluminum 7075 (Al7075). The primary objective is to reduce wear rates and enhance mechanical performance through controlled reinforcement, and to optimize processing and testing parameters using statistical and machine learning-based techniques. Al7075/TiC composites with varying TiC content (0–4 wt%) were fabricated using the powder metallurgy route, which involved mechanical mixing, cold compaction, and sintering to achieve homogeneous particle distribution and high interfacial integrity. The fabricated composites were evaluated for their microstructural, mechanical, and tribological properties. Experiments were conducted under various applied loads (10–30 N) and sliding speeds (1–3 m/s) to investigate the wear behavior. Microstructural analysis confirmed uniform dispersion of TiC particles and strong interfacial bonding with the matrix. Results indicated that increasing TiC content improved composite density (up to 61.42%) and hardness (up to 86.95%). Response Surface Methodology (RSM) was applied in conjunction with advanced optimization strategies inspired by deep learning, including Grid Search, Random Search, and Bayesian Optimization, to identify the optimal parameters for minimizing wear rate. All techniques consistently identified the optimal condition as 4 wt% TiC, 10 N load, and 3 m/s sliding speed, yielding the lowest wear rate of 0.00214 mm3/min. Analysis of variance revealed that TiC content had the most significant effect on wear rate, followed by applied load and sliding speed. Among the optimization methods, Bayesian Optimization demonstrated superior efficiency by effectively exploring the parameter space, making it particularly advantageous for studies constrained by limited experimental data.